NASA Technical Reports Server (NTRS) 20020039333: Functionalization of Carbon Nanotubes using Atomic Hydrogen

10 Pages·0.42 MB·English

by  NASA Technical Reports Server (NTRS)

Functionalization of Carbon Nanotubes
Using Atomic Hydrogen
Bishun N. Khare "'_,Alan M. CasselP, Cattien V. Nguyen _, M. Meyyappan, Jie Han:
NASA Ames Research Center, Moffett Field, CA 94035
tSETI Institute, =EIoret Corporation
ABSTRACT
We have investigated the irradiation of multi walled and single walled carbon
nanotubes (SWNTs) with atomic hydrogen. After irradiating the SWNT sample, a
band at 2940 cm "1(3.4 ltm) that is characteristic of th_ C-H stretching mode is
observed using Fourier transform infrared (FTIR) spectroscopy. Additional
confirmation of SWNT functionalizatlon is tested by irradiating with atomic
deuterium. A weak band in the region 1940 cm "=(5.2/Jm) to 2450 cm "=(4.1 pm)
corresponding to C-D stretching mode is also observed in the FTIR spectrum. This
technique provides a clean gas phase process for the functionalization of SWNTs,
which could lead to further chemical manipulation and/or the tuning of the
electronic properties of SWNTs for nanodevice applications.
Single-walled carbon nanotubes fore, there is a need for developing
(SWNTs) have been considered as single methods for the post-growth
molecular units for nanoelectronic manipulation of the electrical transport
devices, possible sorbents for hydrogen properties of the nanotubes. For
fuel storage, and in numerous other instance, introduction of C-H bonds in
applications t. In order to realize many of metallic SWNTs will introduce barriers
these applications, chemical processing to electron transport that may lead to
methods need to be developed that allow useful semiconducting devices,4. Also,
for the direct manipulation and tuning of hydrogenated nanotubes may impart
the chemical and physical properties. radiation shielding properties to
Gas phase functionalization techniques composites for space applications. Here
could prove extremely useful to fabricate we describe a clean gas phase process
sensors and devices based on SWNTs :'3. for the functionalization of SWNTs
One of the challenges in fabricating producing a C-H bond.
large arrays of SWNT devices is the lack Previous work on multiwall
of selectivity in obtaining SWNTs with carbon nanotubes (MWNTs) has shown
the desired electronic properties. There- one broad asymmetric line at 1575 cm j
and a weaker line at 868 cm l in the
" Email: [email protected] transmission infrared spectrum.
Kuhlmann et al. reported _ weak IR
absorptionsat 1598±3 cm" and 874±2 plasma enhanced gas phase approach. In
cm t in SWNT and 1590 cm z and 868 the cold plasma, the gas flow is
cm" in polycrystalline graphite samples. maintained at a fraction of a Torr
The intensity of the band at 868 cm" is pressure in the plasma chamber. The
-0.015 absorbance unit or 96.6% energy is supplied to the flowing gas by
transmission. Since many of these bands a radiofrequency (RF) generator
are weak, care must be taken to ensure designed to provide up to 100 Watts of
that the collection of the spectrum is not continuous wave (13.56 MHz) power to
masked by background interferences 6. the reaction chamber. Maximum power
Kastner et al. 7 were able to observe the transfer from the RF amplifier tubes to
weak band at 1575 cm" because they the reaction chamber is accomplished by
obtained their spectrum using a vacuum matching the output impedance of the
spectrometer where the background generator to the input impedance of the
absorption due to water vapor is reaction chamber. A glow discharge with
eliminated.
characteristic high electron energies is
There have been limited reports produced in the reaction chamber. The
of the direct functionalization s'9 of temperature of the gas is typically
SWNTs. One of the inherent difficulties -1000 K (0.05 - 0.1 eV), which is much
is the challenge presented in purifying lower than that of the electrons at a few
and handling the SWNTs. Recently eVs. The glow discharge produces
Koshio et al. reacted SWNTs with electrons by ionization and radicals by
monochlorobenzene (MCB) or poly bond dissociation, thus providing the
(methyl methacrylate) (PMMA) by atomic hydrogen for functionalization.
subjecting SWNTs to ultrasonication% The plasma may also introduce defects
In the latter case, SWNTs were in CNTs which will serve to reduce the
dispersed in a 2 % solution of PMMA in activation energy for chemisorption.
MCB. FTIR spectra for SWNT/MCB Our initial experiments employed
mixture produced peaks at 2848 and CVD-grown MWNTs deposited on an
2920 cm" corresponding to the iridium-coated silicon wafer':. No
stretching mode of C-H bond vibration spectral features were found above the
in saturated hydrocarbons. For the background of the FTIR spectrometer.
SWNT/PMMA/MCB mixture, they A Thermo Nicolet Nexus 670 FTIR
found two additional peaks at 2949 and spectrometer was employed in our
2989 cm" suggesting that two different investigations. Reaction of hydrogen
species produced by the decomposition plasma with MWNTs was carried out for
of PMMA reacted with the reactive sites 45 minutes and a small IR signal above
induced on SWNTs by ultrasound. They the background in the C-H stretch region
also found a band at 1729 cm" due to is barely indicated (Figure 1). Further
C=O bond in carbonyl groups that arises plasma processing for a total of 135
from the presence of dissolved oxygen in minutes resulted in no significant change
the solvent. Recently Chen et al. used in the IR band due to limited penetration
plasma chemical modification of CNTs depth of the charged particles. The
and subsequent reactions exploiting penetration depth of electrons with
plasma induced surface functionalities '_. energy 10 eV is only 10 - 20 nm.
Our attempts to functionalize Furthermore, the H atom may possess
carbon nanotubes (CNTs) use a cold much less energy and will therefore
barely processthe outer layers of the metastable species is well established.
MWNTs. This may be the reason why They have extensively been adopted for
the intensity of the C-H band does not experiments with atomic hydrogen in
increase regardless of how long MWNTs several research fields. A schematic of
are irradiated by H atoms in the plasma our apparatus is shown in Figure 2. The
chamber. Additionally, since only the source used in our experiment consists
outer most layer is available for of a Pyrex tube with an inner diameter of
functionalization, the total accessible C 10 mm and 1 mm thickness inserted in
for functionalization is very low an McCarroll cavity operated at 2.45
compared to the total amount of C in the GHz. Microwave power is supplied by
MWNTs.
an Opthos microwave generator (model
It is important to note that while MPG 4M). The input wattage for the
the H2 plasma is on, CNTs are exposed microwave generator was 45 Watts and
to harmful vacuum UV radiation the reflected power was at ~ 8 Watts.
CLyman-a) produced in the plasma that The exit end of the tube is connected to a
can dissociate some of the C-H bonds as
vacuum )chamber (in which the sample is
they are formed. Since we have no located) through a Teflon plug with a 1
control over the penetration depth of the mm bore down the center that collimates
H atoms or the amount of UV radiation the H atom stream toward the target.
in this experimental set-up, we designed This aperture allows the buildup of gas
a new apparatus to eliminate exposure of needed for a localized discharge _7. To
the CNT sample to UV radiation while prevent UV radiation produced in the
permitting H atoms to constantly discharge region from hitting the sample,
impinge on the sample. This setup and the entrance hole of the Teflon plug is
the results will be described further in designed off-axis, which allows only H
detail shortly. atoms to pass.
It is known that UV irradiation We used a special glass cryostat
decreases the intensity of the C-H where a calcium fluoride window coated
stretching feature in hydrogenated with purified SWNTs is fastened in the
carbon materials x3t5 and therefore it is center of the cryostat that is on the end
important to ensure that the rate of C-H of a Dewar container that can be filled
bond formation is higher than the rate of with liquid nitrogen. However, all of our
destruction of the C-H bond by the UV functionalization studies were carried
radiation. Mennella et al. have removed out at room temperature. The sample can
the UV radiation and shown that C-H be rotated to any position as the bottom
modes are activated in nano-sized shroud with two CaF, windows is fixed
hydrogen-free carbon particles by and the top cryostat and the shroud are
exposure to atomic hydrogen _. We have connected by a ground glass joint
adopted this approach for formation of (Figure 2). This allows for recording of
C-H bonds in SWNTs.
the spectra during the irradiation process
Based on Mennella's work j6, we in order to monitor the progress of
wanted to eliminate the UV component functionalization. The bottom shroud has
of the discharge. The application of two openings; one connected to the H
microwave and radio frequency excited atom source and one across connected to
discharge sources for the production of the high vacuum pumping system. At
intense beams of atomic, radical, and 90° to these openings, the shroud is
3
equipped with two CaF, windows order to deposit a film of SWNTs on the
capableof permittinganIRbeamtopass calcium fluoride windows.
through the sample for taking the Figure 3 clearly demonstrates
spectrum. Before we started the that the C-H bonds are formed. The
irradiation, we took a spectrumof the important result is that H atom
SWNTthatservedasthebackground. irradiation on SWNTs produced a band
The pressureof H.,gas at the at 2940 cm j (3.4 lam) with subfeatures
entranceof theUV lampwasmaintained at 2949, 2918, and 2848 cm" typical of
at500pm Hg,while thepressureatthe
the C-H stretching vibrations in CH_, and
exit of the shroud read about 1 p.m Hg. CH 3 groups. There is no evidence for
Since the pressure difference is 500 I,tm aromatic C-H modes (-3040 cmt). We
to 1 p.m, there is a beam of H atoms are not able to see the corresponding
hitting the SWNTs on CaF, substrate. bending mode because it is obscured by
HiPCO derived SWNT samples a strong water band from the background
provided by Rice University _swere used of the spectrometer. It is noted that the
in this study. Purification of the SWNTs intensity_ of the peaks in Fig. 3 does not
was carried out by transferring 50 mg of change regardless of the irradiation time
the sampie to a 50 ml flask with the from 1 to 7 hours. Future controlled
addition of 25 ml of concentrated HCf experiments will determine the
and I0 m[ of concentrated HNO_. The minimum irridation time required for
solution was heated for 3 hours and functionalization and extent of hydrogen
constantly stirred with a magnetic stirrer coverage as these are all issues of
in a reflux apparatus equipped with a importance in applications. To confirm
water-cooled condenser. This was done C-H stretching mode in SWNTs, we
to remove the unwanted iron/graphite functionalized with D atoms by
nanocrystallites that are a by-product of replacing hydrogen gas with deuterium
the production process. The resulting in the microwave discharge. A weak
suspension was then transferred into band in the region 1940 cm "j (5.2 ,um) to
centrifuge tubes and spun-down at 3220 2450 cm "t (4.1 lam) corresponding to
g for 30 minutes. After pouring off the C-D stretching mode is observed in the
supernatant, the solid was resuspended FTIR spectrum. However, the
and spun-down (30 min) three times in background spectrum of air (CO,. and
deionized water. Next, the solid was H20) in the spectrometer combined with
treated with NaOH (0.01 M) and the limited signal intensity prevents us
centrifuged for 30 minutes. We from confirming this band. We are
qualitatively confirmed the purity of the currently developing the use of a
SWNTs before and after purification vacuum FTIR spectrometer so we have
using transmission electron microscopy no subtraction problem of the CO_, and
(data not shown here). As evidenced by H,.O bands.
TEM, removal of the Fe nanopanicles Direct hydrogenation of SWNT
was accomplished, and the crystal]inity has been studied by quantum chemistry
and other theoretical methods "J with the
of the SWNTs was maintained. Finally,
conclusion that the addition of molecular
the sample was dried using a vacuum
oven kept at 60 ° C overnight. The hydrogen to nanotube is a slightly
SWNTs were then suspended in CCI.t in exothermic process. Atomic hydrogen
can chemisorb on CNT sidewalls with a
4
bindingenergyof -26 kcal/mol:". The short alkyl groups attached to the
chemisorption and desorption energies sidewalls at these openings. However,
of H on CNT, however, depend on the there is no clear evidence for such
coverage of H on the nanotube. At low substitution reaction here.
coverages, the binding energy of an In conclusion, we have
isolated H might be little lower than successfully functionalized SWNTs
when other H atoms start to chemisorb using hydrogen atoms from a microwave
on the neighboring sites. At very high discharge. A key parameter was the
coverages however, H atoms from the elimination of the UV component from
neighboring sites could spontaneously the discharge. This clean and simple
desorb as H: molecules. We expect, technique has the advantage of giving
therefore, the system to be in a dynamic very clear results compared to wet
equilibrium at an intermediate coverage chemical reactions which can produce
between the above two limits. The H polymeric material that produce noise in
chemisorption coverage and the FTIR the transmission spectra. The approach
line shapes and positions will not change can be qsed to functionalize nanotubes
much after that limiting coverage has with other species such as N, F, NH, etc.
been achieved. An indication of the Acknowledgements are due to
above scenario is that in the experiments Rice University CNST for providing
we do not observe any change in the SWNTs, Raj Khanna for providing the
spectra when the exposure time is glass Cryostat, Bruce Borchers for
increased from about 30 minutes to few drawing the apparatus and machining the
hours. More accurate time resolved teflon plug, Sunita Verma, Metages
data, in future, for exposure times much Sisay, Hiroshi Imanaka, Matthew
shorter than 30 minutes may help to Cannady, Dave Scimeca, and Thomas
verify the above explanation. Based on Frey for assistance with the experiment.
mechanisms proposed by Kuznetsova et Discussions with V. Mennella, Deepak
al', it is likely that plasma processing of Srivastava, Charlie Baushclicher, Harry
the SWNT sample creates openings in Partridge, Ramsey Stevens and Jun Li
the sidewalls with dangling-C=C are acknowledged. Chris McKay, Dale
groups which are susceptible to Cruikshank, Yvonne Pendleton and
hydrogenation. In this manner the Michael Flynn are acknowledged for
functionalized nanotubes probably have supporting BK's laboratory facilities.
References
(1) Kuznetsova, A.; Mawhinney, D.B.; Naumenko, V.; Yates, Jr. J. T.; Liu, J.; Smalley
R. E. Chem. Phys. Lett. 2001, 321,292.
(2) Kong, J.; Franklin, N. R.: Zhou C.; Chapline, M. G.; Peng, S.; Cho, K.; Dai H.
Science, 2000, 287, 622.
(3) Collins, P. G.: Bradley, K.; Ishigami, M.; Zettl A. Science, 2000, 287, 1801.
(4) Liu, Y.; Jaffe, R.: Han, J. Phys. Rev. B, 2001, to be submitted.
(5) Kuhlmann, U.; Jantoljak H.; Pfander, N.; Bernier, P.; Journet, C.: Thomsen, C.
Chem. Phys. Lett. 1998, 294, 237.
(6) The other asymmetric band at 1575 cm j is masked by a very strong O-H bending
°
mode of the water molecule in our spectrometer. After purging the spectrometer
with dry air and subtracting the residual water remaining after purging, it is
difficult to justify its presence with certainty because to start with, this [R
active CNT band is also a weak line (0.02 absorbance unit or 95% transmission).
(7) Kastner. J.; Pichler, T.; Kuzmany, H.; Curran, S.; Blau, W.; Weldon, D. N.;
Delamesiere, M.; Draper, S.; Zandbergen, H. Chem. Phys. Left. 1994, 221, 53.
(8) Mickelson, E. T.; Huffman, C. B.; Rinzler, A. G.; Smalley, R. E.; Hauge, R. H.;
Margrave, J. L. Chem. Phys. Lett. 1998, 296, 188.
(9) Hamwi, A.; Alvergnat, H.; Bonnamy, S; Begun, F. Carbon 1997, 35, 72.
(I0) Koshio, A.; Yudassaka, M.; Zhang, M.; lijima, S. Nano Iytt. 2001 XXXX, A-C.
(1 I) Chen, Q.: Dai, L.; Gao, M.; Huang S.; Mau A. J. Phys. Chem. B 2001, 105,
618.
(12) Cassell, A. M.; Verma, S.; Delzeit, L.; Meyyappan, M.; Han J. Langmuir 20@I,
17, 260.
(13) Iida, S40htaki, T.; Seki, T. in AlP Conf. Prec. 120 Optical Effects in Amorphous
Semiconductors, ed. P. C. Taylor and S. G. Bishop (l_/e.w York: AlP) 1984, 258.
(14) Munoz, G. M.; Ruiterkamp, R.; Schutte, W. A.; Green_rg, J. M.; Mennella, V.
Astronomy and Astrophysics 20@1,367, 347•
(15) Menneila, V.; Munoz Caro, G. M.; Ruiterkamp, R.; Shutte, W. A.; Cn'eenberg, J.
M.; Brucato, J. R.; Colangeli, L. Astronomy and Astrophysics 2001, 367, 355.
(16) Mennella, V.; Brucato, J. R.; Colangeli, L. Astrophys. J. 1999, 524, L71.
(17) Paolini, B. P.; Khakoo, M. A. Rev. Sci. Instrum. 1998, 69, 3132.
(18) Nikolaov, P.; Bronikowski, M. J.; Bradley, R. K.; Rohmund, F.; Colbert, D.T.;
Smith, K. A.; Smailey, R. E. Chem. Phys. Lett. 1999, 313, 91.
(19) Bauschlicher, Jr., C. W. Nano Lett. 2001, 1 223.
(20) Jaffe R.; Srivastava D. (unpublished).
6
Figure Captions:
Figurel. FTIR spectra of MWNTs after processing with cold hydrogen plasma. C-H
stretch above the noise level is indicated. Curve (a) is after 45 rain irradiation and curve
(b) is after 135 rain.
Figure2.Apparatus toirradiatSeWNTs by H atoms produced inamicrowave discharge.
H atoms enterthrough two Imm holes(a)inaTeflonplugat90°tothetarget.This hole
isjoinedby another Imm hole(b)throughthecenteroftheplug directingtheH atoms in
a beam tothetarget.
Figure 3. FTIR spectra of SWNTs irradiated by H-atoms. C-H bands at 2949, 2918, and
2848 cm" are clearly indicated.
! -,
7
5
.(a)
C:
E
(b)
I,--
I--
o_
, , , , I , , , , I , , , , I , , , , I , , , , I , , , , l J , , , I , J , ,
3300 3200 3100 3000 2900 ' 2800 2700 2600 2500
Wavenumbers (cm-1)
Figure 1.
\
I
I
I
45/50 Taper
Pyrex Joint
Hydrogen In
CaF2 Window
(a) McCarrol Cavity
(b)
H Atoms
To Vacuum
Sample
Figure 2.
5
V
(D
(.)
(-
E
O9
(.-
I,,.,
I--
i , i , | i J i , I , ¢ i i | . , i i | , ,
3300
3200 3100 3000 2900 '28'00 2700 2600 2500
Wavenumbers (cm-1)
Figure 3.
10